Table 2. Calculated emission rate (in megagrams per day) ofsome volatile elements from Mount Erebus in December 1986

and estimated emission rates in 1983 assuming the sameelement-to-sulfur ratios as in December 1986.

Element 1986 1983

Fluorine 41.8 ± 4.1 484Sodium 6.8 ± 1.6 78Chloride 107.0 ± 0.7 1,234Potassium 5.9 ± 1.8 68Copper 0.2 ± 0.1Zinc 0.25 ± 0.1Arsenic 0.09 ± 0.03Bromine 0.24 ± 0.05

NOTE: Calculations used the measured sulfur dioxide emission rates of20 and 230 megagrams per day in 1986 and 1983, respectively.

dances of the elements in the phonolite magma. Indium is themost enriched element in the volcanic plume relative to itsconcentration in the magma and is followed by chlorine, flu-orine, arsenic, and sulfur.

Absolute emission rates for selected elements have beendetermined using the COSPEC 2 sulfur dioxide data and theelement/sulfur ratios determined from the filter packs (table2). In December 1986, the emission rates of hydrogen chlorideand hydrogen fluoride were 107 and 42 megagrams per day,respectively. These emission rates are significantly higher thansulfur dioxide which commonly dominates over hydrogenchloride and hydrogen fluoride at most other volcanoes aroundthe world. This is a function of the evolved nature of thestrongly undersaturated alkalic phonolitic magma. If the com-position of the gas plume remains unchanged with time, thenthe COSPEC results from December 1983 can be combinedwith the 1986 treated filter data to give an estimate of the

hydrogen chloride and hydrogen fluoride emissions rates of1,234 and 484 megagrams per day, respectively (table 2). Theseestimated emission rates for 1983 are extremely high and sug-gest that Mount Erebus has been an important source of aero-sols to the antarctic atmosphere.

This work was supported by National Science Foundationgrant DPP 85-19122. We appreciate the assistance of the VXE-6 helicopter crews and ITT Antarctic Services personnel forlogistic support.

References

Finnegan, DL., J.P. Kotra, D.M. Hermann, and W.H. ZoIler. 1989.The use of 7LiOl-l-impregnated filters for the collection of acidic gasesand analysis by instrumental neutron activation analysis. Bulletin ofVolcanology, 51, 83-87.

Germani, M.S. 1980. Selected studies of four high-temperature air pollutionsources. (Doctoral thesis, University of Maryland, College Park,Maryland.)

Kitto, ME., D.L. Anderson, and W.H. Zoller. 1988. Simultaneouscollection of particles and gases followed by multielement analysisusing nuclear techniques. Journal of Atmospheric Chenistry, 7, 241-259.

Meeker, K. 1988. The emission of gases and aerosols from Mount Erebusvolcano, Antarctica. (Master of science thesis, New Mexico Instituteof Mining and Technology, Socorro, New Mexico.)

Rose, WI., R.L. Chuan, and P.R. Kyle. 1985. Rate of sulphur dioxideemission from Erebus volcano, Antarctica, December 1983. Nature,316, 710-712.

Stoiber, RE., L.L. Malinconico, and S.N. Williams. 1983. Use of thecorrelation spectrometer at volcanoes. In H. Tazieff, J.C. Sabroux,(Eds.), Forecasting volcanic events. Amsterdam: Elsevier.

Symonds, RB., P.R. Kyle, and W.I. Rose. 1985. SO 2 emission ratesand the 1984 activity at Mount Erebus Volcano, Antarctica. EOS, 66,417.

Taylor, S.R. 1964. Abundance of chemical elements in the continentalcrust: A new table. Geochimica et Cosmochimica Acta, 28, 1,273-1,285.

The dry valleysmafic dike swarm,

southern Victoria Land

WALTER R. VENNUM*

U.S. Geological SurveyFederal Center

Denver, Colorado 80225

The dry valleys lie in the Royal Society Range of the Trans-antarctic Mountains in southern Victoria Land. The oldest ex-posed rocks in this area are the metasedimentary Skeleton

* Present address: Department of Geology, Sonoma State University, Roh-nert Park, California 94928.

Group of Cambrian or Late Proterozoic age. The Skeleton Groupis intruded by a suite of locally deformed granitic to dioriticrocks (the Granite Harbor Intrusive Suite) which was emplacedduring the Ordovician Ross orogeny. Numerous series of maficto felsic late synorogenic dikes also of probable Ordovician age(Berg 1988) cut this basement complex in many parts of south-ern Victoria Land. Gunn and Warrer (1962) have described thegeology of southern Victoria Land and Keiller (1988) has dis-cussed the complex field relations of the dikes.

The synorogenic dikes include as many as four cross-cuttingsets of dark, mafic, aphanitic dikes (Angino, Turner, and Zeller1962). Numerous geologists have referred to these dikes aslamprophyres. These dark, mafic, aphanitic dikes, however,do not contain euhedral mafic phenocrysts and thus do not fitthe classical definition of lamprophyres (Streckeisen 1979; Rock1984). An extensive swarm of these dikes is well exposed ineastern Wright, Taylor, and Victoria dry valleys. The field re-lations of this dike swarm, henceforth referred to as the dryvalleys mafic dike swarm, were examined and samples were col-lected at five localities (Nussbaum Riegel, Mount Insel, MountCerebus, the Eastern shore of Lake Vanda, and on the ridge

1990 REVIEW 11

crest west of the Meserve Glacier in Wright Valley). This articleclassifies these dikes by describing their mineralogy and pe-trography and presents data of a reconnaissance geochemicalnature. These results are then compared to similar data derivedfrom caic-alkaline lamprophyres elsewhere in the world.

The mineralogy of all dikes is similar. Approximately 60percent of each dike is composed of felsic minerals, largelyplagioclase. Most plagioclase crystals are andesine or oligo-clase, occur as moderately zoned tablets or laths and are weaklyto moderately altered to a mixture of sericite and/or variousclay minerals. Amphibole is typically the most abundant maficmineral and occurs as euhedral to subhedral prismatic crystalsthat are usually unaltered. Its pleochroic formula and absorp-tion scheme are: Z = orange-brown, Y = pale orange-brown,X = pale yellow; Z>Y>X. Some amphibole crystals have thinrims of green actinolite (?). Biotite is present in all samples;locally it is more abundant than amphibole or present only inaccessory amounts. Its pleochroic formula and absorptionscheme are Z = Y = dark red to red black, X = pale yellow;Z = Y>X. Clinopyroxene grains, most of which are stronglyaltered to mixtures of epidote and sericite, present in almostall samples, but only in accessory amounts.

The groundmass contains abundant minute apatite needles,minor quartz and carbonate, and sparse potassium feldspar.Ocelli (globular structure) of epidote or of quartz rimmed withsericite occur in about 30 percent of the thin sections. Twotypes of chlorite occur interstitially in the groundmass or re-place biotite. The overall groundmass texture is equigranular,with individual grains ranging between 0.25 and 1.5 milli-meters in diameter. Microphenocrysts of amphibole, biotite,

and less commonly plagioclase as much as 3 to 4 millimetersin diameter occur sparsely in about half of the thin sections.

Electron microprobe analyses of the mafic minerals indicatethe amphiboles are magnesian hastingsite or ferroan pargasite(Leake 1978), the pyroxenes are magnesium-rich augite (Mor-imoto 1988) and the micas are biotites containing subequalamounts of the siderophyllite and annite molecules (Bailey1984). Overall mineralogy and specific composition of the am-phiboles and pyroxenes of the dry valleys mafic dike swarmare similar to the mineralogy and mafic mineral compositionsdescribed by Rock (1984, 1987) from spessartite lamprophyres(a subtype of calc-alkaline lamprophyres composed dominatelyof amphibole and plagioclase) elsewhere in the world. Biotitesin the dry valley mafic dike swarm have much higher iron-to-magnesium ratios than micas which generally occur in calc-alkaline lamprophyres (Rock 1984, 1987).

Major, minor, and trace-element contents of 10 whole-rocksamples and rare-earth element compositions of three of thesesamples are reported in the table where they are also comparedwith averages of 129 worldwide spessartile lamprophyre anal-yses (Rock 1984). Dry valleys mafic dike swarm samples av-erage 2.5 percent more silicon-oxide and 0.25-1.5 percent lessferric oxide, magnesium oxide, and sodium oxide than Rock's(1984) average spessartite lamprophyre. Values for all reportedmajor and minor element oxides, however, fall within the geo-chemical screens used by Rock (1984) to discriminate caic-al-kaline lamprophyres. The volatile contents, especially carbondioxide, of the dry valley mafic dike swarm samples are con-siderably lbwer than average spessartite lamprophyre and ap-proach those of "common mafic igneous rocks" (LeMaitre 1976).

Geochemical data from the dry valleys mafic dike swarm compared with average composition of worldwide calc-alkaline lamprophyres

A B A Ca

Silicon oxideAluminum oxideFerric oxideFerrous oxideMagnesium oxideCalcium oxideSodium oxidePotassium oxideTitanium oxidePhosphorus

oxideMaganese oxideStructural waterAdsorbed waterCarbon dioxide

54.5 (51.9-58.4)15.6 (14.5-17.3)1.57 (1.24-2.23)5.84 (5.24-6.88)5.66 (2.49-7.02)7.33 (5.75-8.40)2.92 (2.33-3.38)2.36 (1.09-3.40)1.04 (0.84-1.23)

0.31 (0.21-0.43)0.14 (0.11-0.19)2.02 (1.32-2.52)0.15 (0.07-0.35)0.12 (0.02-0.46)

51.9Barium 794 (400-1,000)15.1Beryllium 2 (1-3)2.9Cobalt 28 (15-33)5.4Chromium 281 (6.450)6.8Copper 21(3-63)7.0Gallium 23 (21-25)3.2Lithium 39 (25-73)2.4Niobium 15 (12-17)1.3Nickel 83(2-170)

0.4Lead 14 (5-26)0.1Rubidium 85 (50-130)2.4Scandium 18(11-26)

_bStrontium 615 (550-690)1.1Thorium 6 (4-10)

Vanadium 109 (67-140)Yttrium 26 (22-30)Zinc 196 (130-225)Zirconium 196 (130-225)

LanthanumCeriumPraseodymiumNeodymiumSamariumEuropiumGadoliniumTerbiumDysprosium

HolmiumErbiumThuliumYtterbiumLutetiumNormalized rare-earth ratioEuropium anomalyY.14 rare-earth elements

33.0 (24.9-39.6)66.5 (49.9-79.2)

7.8(5.8-9.2)

33.7(27.5-38.1)

6.7(5.7-7.3)

1.81(1.68-1.90)

6.9(6.4-7.5)

1.0(0.9-1.2)

5.1(4.8-5.4)

0.97 (0.84-1.05)

2.7(2.4-3.0)

0.36(0.31-0.42)

2.4(2.0-2.8)0.35 (0.30-0.42)

7.4(4.5-10.1)0.82 (0.76-0.85)

170(136-194)

a Column heads: A. Average and range of 10 dry valley dike analyses. B. Average of 129 worldwide spessartite lamprophyres (Rock 1984). C. Averageand range of three dry valley dike analyses.

b Not reported.NOTE: Rapid rock major-element analyses by J. Ardith, J. Bartel, E. Brandt, S. Roof, K. Stewart, and J. Taggart of the U.S. Geological Survey.Methods used are described by Shapiro and Bannock (1962) and are supplemented by atomic absorption. X-ray fluorescence trace element analysesby C. Gent and B. King of the U.S. Geological Survey, Inductively coupled plasma emission spectroscopic rare-earth element analyses by J. crockand K. Kennedy of the U.S. Geological Survey. Major elements in weight percent; trace and rare-earth elements in parts per million. Not detected orat limit of determination for all McMurdo Dry Valley samples: silver, arsenic, gold, boron, bismuth, cadmium, germanium, molybdenu, tin, tantalum,uranium, and tungsten.

12 ANTARCTIC JOURNAL

cc000cc

U!

500

U)

C0C0wwcc

In the JUGS total alkali-silicon oxide classification of volcanicrocks (Le Bas et al. 1986) five of the analyses are basaltic an-desites, three are potassic trachyandesites and two are andes-ites. In Macdonald and Katsura's (1964) classification the threetrachyandesites lie in the alkaline field whereas the remainingseven analyses are tholeiitic. All dry valleys mafic dike swarmanalyses plot in Rock's (1987) caic-alkaline lamprophyres field.CIPW norms (not listed in the table) were calculated using theferric oxide-to-ferrous oxide ratio (0.18) of the least oxidizedsample. Two of the trachyandesites are olivine-diopside-hy-persthene normative, the other eight analyses are quartz-diop-side-hyperthene normative. There are no appreciablegeochemical or mineralogical differences or trends apparentbetween younger and older dikes, between margins and in-teriors of individual analyzed dikes or between dikes fromdifferent locations.

Average trace-element contents of the dry valleys mafic dikeswarm samples fall within the ranges quoted by Rock (1987)for calc-alkaline lamprophyres, but these ranges are so largeas to be almost meaningless. A much better way of comparingtrace-element data is to examine a spidergram (figure 1) ofmid-ocean ridge basalt normalized elemental abundances. Al-though the dry valley mafic dike swarm samples contain some-

Sr K Rb Ba Th Nb Ce P Zr Hf Sm Ti Y Yb Sc Cr

Figure 1. MORB-normalized spidergram comparing average distri-bution of various elements in dry valleys mafic dike swarm sampleswith those of average worldwide calc-alkaline lamprophyres (Rock1984). Normalizing values are from Pearce (1982). (Chemical sym-bols are: Sr, strontium; K, potassium; Rb, rubidium; Ba, barium;Th, thorium; Nb, niobium; Ce, cerium; P, phosphorus; Zr, zircon-ium; Hf, hafnium; Sm, samarium; Ti, titanium; Y, yttrium; Yb, yt-terbium; Sc, scandium; and Cr, chromium.)

La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Figure 2. Average chondrite-normalized rare-earth element patternfor dry valleys mafic dike swarm samples compared to averagechondrite-normalized rare-earth element pattern for worldwide calc-alkaline lamprophyres. Latter data modified from Rock (1987, figure8). Normalizing values are from Nakamura (1974). (Chemical sym-bols are: La, lanthanum; Ce, cerium; Pr, praseodymium; Nd, neo-dymium; Pm, promethium; Sm, samarium; Eu, europium; Gd,gadolinium; Tb, terbium; Dy, dysprosium; Ho, holmium; Er, erbium;Tm, thulium; Yb, ytterbium; and Lu, lutetium.)

what less barium, cerium, and thorium than average calc-alkalinelamprophyre, the two trends on the spidergram are very sim-ilar to each other including the characteristic niobium deple-tion. Figure 2 compares chondrite-normalized rare-earth elementprofiles for worldwide calc-alkaline lamprophyres with similardata from the dry valleys mafic dike swarm. These sampleshave several times (3-5) less total normalized light rare-earthelements than average calc-alkaline lamprophyre and also dis-play small (0.82) negative europium anomalies, whereas av-erage calc-alkaline lamprophyre do not have europiumanomalies. The two trends are otherwise similar in their overallslope and pattern.

Geochemistry, mineralogy, and composition of specific maficminerals from the dry valleys mafic dike swarm are with fewexceptions, similar to spessartite-type calc-alkaline lampro-phyre from the rest of the world. The dry valleys mafic dikeswarm crystallized from a volatile poor, silicon-oxide enrichedcalc-alkaline lamprophyre magma with a lower-than-averagecontent of ferric oxide, magnesium oxide, sodium oxide, bar-ium, cerium, thorium, and light rare-earth elements. The dryvalleys mafic dike swarm does not possess one of the char-acteristic features of lamprophyres (euhedral mafic pheno-crysts) except on a microscopic basis and are best referred toas "microlamprophyres."

1990 REVIEW 13

This research was supported by National Science foundationgrant DPP 83-18183 to the U.S. Geological Survey. A previousversion of this manuscript benefited from reviews by P.D.Rowley and M.A. Kuntz of the U.S. Geological Survey.

References

Angino, E.E., M.D. Turner, and E.J. Zeller. 1962. Reconnaissancegeology of lower Taylor Valley, Victoria Land, Antarctica. GeologicalSociety of America Bulletin, 73, 1,553-1,562.

Bailey, S.W. 1984. Classification and structures of the micas. In S.W.Bailey (Ed.), Micas (Reviews in mineralogy, vol. 13). Washington,D.C.: Mineralogical Society of America.

Berg, J.H. 1988. Mafic and ultramafic inclusions in lamprophyre dikesfrom the Royal Society Range of the Transantarctic Mountains. Ant-arctic Journal of the U.S., 23(5), 30-31.

Gunn, B.M., and C. Warren. 1962. Geology of the basement rocksbetween the Mawson and the Mulock Glaciers, South Victoria Land,Antarctica. New Zealand Geological Survey Bulletin, 71, 1-157.

Keiller, I.G. 1988. Comments and observations on dyke intrusions ofthe Wright Valley. New Zealand Antarctic Record, 8(3), 25-33.

Leake, B.E. 1978. Nomenclature of amphiboles. American Mineralogist,63, 1,023-1,052.

Le Bas, M.J., R.W. Le Maitre, A. Streckeisen, and B. Zanettin. 1986.A chemical classification of volcanic rocks based on the total alkali-silica diagram. Journal of Petrology, 27, 745-750.

Le Maitre, R.W. 1976. The chemical variability of some common ig-neous rocks. Journal of Petrology, 17, 589-637.

Macdonald, GA., and T. Katsura. 1964. Chemical composition of Ha-waiian lavas. Journal of Petrology, 5, 82-133.

Morimoto, N. 1988. Nomenclature of pyroxenes. American Mineralogist,73, 1,123-1,133.

Nakamura, N. 1974. Determination of REE, Ba, Mg, Na and K incarbonaceous and ordinary chondrites. Geochemica et CosmochimicaActa, 38, 757-775.

Pearce, J.A. 1982. Trace element characteristics of lavas from destruc-tive plate boundaries. In R.S. Thorpe (Ed.), Andesites: Orogenic an-desites and related rocks. New York: John Wiley and Sons.

Rock, N.M.S. 1984. Nature and origin of calc-alkaline lamprophyres:Minettes, vogesites, kersantites and spessartites. Transactions of theRoyal Society of Edinburgh: Earth Sciences, 74, 193-227.

Rock, N.M.S. 1987. The nature and origin of lamprophyres: An over-view. In J.G. Fitton and B.G.J. Upton (Eds.), Alkaline igneous rocks(Geological Society Special Publication No. 30). London: Blackwell.

Shapiro, L., and W.W. Brannock. 1962. Rapid analysis of silicate,carbonate, and phosphate rocks. U.S. Geological Survey Bulletin, 1,144-A, 1-56.

Streckeisen, A. 1979. Classification and nomenclature of volcanic rocks,lamprophyres, carbonatities and melilitic rocks. Geology, 7, 331-335.

Exposure ages and erosion ratesfor bedrock surface samples

from antarctic dry valleysand nunataks

ROBERT GIEGENGACK

Geology DepartmentUniversity of Pennsylvania

Philadelphia, Pennsylvania 19104-6316

JEFFREY KLEIN, BARBARA LAWN, DAVID FINK,and Roy MIDDLETON

Physics DepartmentUniversity of Pennsylvania

Philadelphia, Pennsylvania 19104-6396

We have developed a method of studying geologic processesthat occur within the top few meters of the Earth's surface(e.g., Klein et al. 1986). The method is based on the accu-mulation of radioactive nuclides produced by nuclear-spalla-tion reactions induced by cosmic rays on major elementsconstituting common minerals. The production rate of thesenuclides is low, ten to a few hundred atoms per gram per year;their measurement was not possible before the advent of ac-celerator mass spectrometry, which is capable of measuring

less than 1 million atoms at isotopic ratios as low as 10-16(Klein, Middleton, and Tang 1982; Middleton et al. 1983; Mid-dleton and Klein 1986).

Application of the method to problems of geologic interestis based on the reduction in the production rates, of radio-nuclides produced in situ, by the overburden of material thathas separated the sample in which they are measured fromthe rock-air interface over the duration of the mean lives ofthose nuclides. The radionuclides calcium-41, aluminum-26,and beryllium-10 have half-lives that range from 100,000 to 1.5million years, making it possible to study subaerial exposuretimes of a few hundred to several million years. Erosion ratesof 10-2 to iO centimeters per year can be measured. Pro-duction rates are reduced by a factor of two under 120-150grams per cubic centimeter of overlying material, equivalentto 30-40 centimeters of rock or about 1.5 meters of ice.

Measurements of more than one radioactive product or ofa radioactive product and a stable one (such as helium-3 orneon-21) often yield information that is qualitatively more valu-able than that learned from measuring only one. The rate ofbuildup of a radionuclide during exposure depends on its pro-duction rate; its decrease during burial depends on its half-life. The study of two or more radionuclides makes it possibleto distinguish single uninterrupted exposures from sequencesof exposures separated by periods of burial. In addition, com-parison of two or more cosmogenic nuclides can be used toeliminate ambiguities of interpretation that are caused by con-tributions from production sources not occurring within therock—specifically atmospheric production. In the case of be-ryllium-10 and aluminum-26, the relative production rates are1:3 in quartz, but 300:1 in the atmosphere; thus, it is possible,

14 ANTARCTIC JOURNAL